Topic: performance

7 stories found

Yesterday

ai_labs75

How enabling two settings tripled our scores on the ARC-AGI-3 benchmark

Enabling two specific API settings significantly boosted GPT-5.6's performance on the ARC-AGI-3 benchmark, tripling scores while improving efficiency by enhancing reasoning capabilities and compacting data retention.

openai.com↗

Monday, July 27, 2026

research40

Adversarial Style Optimization: Enhancing VLM Jailbreaks by GRPO-based Stylistic Triggers Optimization

arxiv.org↗

Saturday, July 25, 2026

newsletters48

[AINews] Claude Opus 5: Fable-level performance at Opus price (half Fable)

latent.space↗

industry24

Designing High-Performance GPU Kernels with TileLang: Tensor-Core GEMM, Fused Softmax, FlashAttention, and Autotuning

marktechpost.com↗

Wednesday, July 22, 2026

research40

Reasoning Fine-Tuning Induces Persistent Latent Policy States

arxiv.org↗

Monday, July 20, 2026

open_source62

affaan-m/ECC — The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development f

github.com↗

research35

Do Coding Agents Need Executable World Models, Simplification, and Verification to Solve ARC-AGI-3?

arxiv.org↗

🌿 That's all for now. Come back tomorrow.

7 of 7 items shown. Sources: 71 days indexed.